http://www.dpreview.com/news/0704/07041601sxsmemorycardformat.asp

Nice ! Here is the replacement path for tape. At some point, we can own a few of these and then back them up to BlueRay or HD-DVD.

lets hope the prices arent a reflection of P2...

This is really interesting. Expensive, but interesting.

This May, the 16 gig P2 cards will be priced at 900 dollars so by the time the SxS comes out, it might be no more than 450 dollars.

Anyway, I’m surprised nobody hinted what the price of this might be.

Why is it P2 cards are still so expensive vs. these new SxS cards that are less?
whats the deal?

I believe it's because when P2 was being engineered, the only way to achieve the necessary performance was by the use of certain techniques and control such as inbuilt raid arrays, which give an inherent fairly high cost per card before the actual memory cost itself is even considered. And this is coupled with relatively low volume production compared to other memory formats.

Nowadays, technology has moved on, and such techniques are no longer necessary for required performance - memory is inherently faster.

I believe it's because when P2 was being engineered, the only way to achieve the necessary performance was by the use of certain techniques and control such as inbuilt raid arrays, which give an inherent fairly high cost per card before the actual memory cost itself is even considered. And this is coupled with relatively low volume production compared to other memory formats.

Nowadays, technology has moved on, and such techniques are no longer necessary for required performance - memory is inherently faster.

The other cost factor was that the memory used in P2 was spec'd as zero defect memory...IOW, the cream of the crop and the suppliers charge a premium price for that stuff. For many years, there were three basic specs for semiconductor devices. Mil Spec, Industrial, and Consumer. The pricing was set accordingly.

-gb-

The other cost factor was that the memory used in P2 was spec'd as zero defect memory...
Yes, and I believe defects in memory don't show up as defects in recorded data (contrary to some popular belief) but the process of isolating and not using such locations makes the chip slower overall. That was significant in the early days of P2 development - now the huge improvement in memory speeds generally mean that for general video usage the cost issues are far more significant.